Supermicro’s X13 family includes a wide range of systems perfect for AI workloads. GPU and SuperBlade systems offer maximum power for training generative AI, large language models and computer vision, while rackmount systems such as Hyper and CloudDC can be configured for AI inferencing from the data center right out to the intelligent edge using GPU acceleration or the built-in accelerators on new 5th Gen Intel Xeon processors.

Supermicro’s complete rack integration services can help our customers build large-scale AI and HPC clusters which are optimized for the best performance and efficiency. We can design, integrate, validate and deliver L11 and L12 rack solutions optimized for our customers’ specific workloads, ready to run from day 1. In addition, Supermicro offers liquid cooling solutions with direct-to-chip liquid cooling for both CPUs and GPUs, which not only allow X13 systems to run at maximum performance, but also help reduce TCO through more efficient cooling.

Supermicro X13 architectures feature optimal compute density, networking flexibility and optional GPU acceleration for the most compute-intensive applications. The density-optimized, highly efficient SuperBlade supports up to 20 nodes in an 8U enclosure with shared power, cooling and ethernet switch, while the single-processor optimized FatTwin is available in 4U 4-node and 4U 8-node configurations.

The new 5th Gen Intel Xeon processors provide better performance per watt at the same power envelope to deliver more processing capacity for compute-intensive HPC applications. Supermicro direct-to-chip liquid cooling solutions can also be utilized to allow higher powered processors in these dense architectures while also further reducing TCO.

The cloud is all about efficiency and scale, and Supermicro’s broad X13 portfolio has a system for any cloud workload. From our powerful, highly flexible Hyper to the cost-effective all-in-one CloudDC, or multi-nodes like BigTwin and GrandTwin for virtualization and content distribution, Supermicro X13 not only makes it easy to customize a system solution for cloud-specific workloads, but also makes it easy to scale as the business grows.

5th Gen Intel Xeon processors are the most efficient Xeon processors ever, with more cores and more performance at the same power envelope as the previous generation. For the first time, Intel Trust Domain Extensions is available on the CPU die, which means hardware-based isolation for virtual machines, designed to protect sensitive data and applications from unauthorized access.

Emerging AI applications can generate massive amounts of data even from a single user, and these workloads require application-specific architectures at every stage of the data pipeline. Supermicro’s flagship X13 Petascale storage platform offers the best architecture to drive large-scale, data-intensive AI and HPC workloads. With up to 24 E3.S drives in 2U and unprecedented end-to end PCIe Gen5 performance, X13 Petascale can help organizations to reach their performance and capacity goals with greater rack density than ever before. For cost-effective scale-out storage, the X13 Simply Double and Front-Loading families provide high density configurations ideal for public & private cloud and data repatriation applications.

The more powerful 5th Gen Intel Xeon processors support the latest PCIe Gen 5 standard to handle the high throughput of a large number of NVMe drives and get the maximum performance out of the new Gen 5 E3.S drives. For large-scale storage, the increased performance of 5th Gen Intel Xeon processors means that a single processor can support some architectures which previously required dual processors.

Efficiency is everything at the edge, and no matter what kind of 5G or Edge workload, Supermicro has an optimized system. The multi-node X13 SuperEdge system is perfect for OpenRAN and private 5G deployments, with front-access expansion slots for accelerator, radio connectivity and retimer cards, while short-depth form factors with front IO are designed to fit into existing telco cabinets.

The new 5th Gen Intel Xeon processors deliver higher performance-per-watt in the same or lower power envelope, resulting in more efficient systems optimized for Edge workloads. With built-in Intel vRAN Boost on the CPU die, Edge workloads can benefit from up to 2x vRAN capacity gains and an additional 20% in compute power savings on vRAN workloads.

Supermicro’s X13 Mainstream and Workstation servers are ideal for small and medium organizations that require flexible, cost-effective systems to run day-to-day business applications. Dual processor architectures deliver ample compute power, supporting a wide range of storage and I/O options for application-specific optimization. Workstation systems bring rackmount power in convenient tower form factors for 3D design, multimedia and enterprise compute applications, ideal for offices, schools and research laboratories.

The 5th Gen Intel Xeon processors include Intel Accelerator Engines to offload workload-specific tasks from the CPU, eliminating the need for discrete accelerator cards and improving overall CPU performance and efficiency. The new processors also feature more cores at the same power envelope, allowing small and medium-sized deployments to run lower-TDP processors or fewer servers, reducing power consumption and cooling infrastructure requirements.